Date: Thu, 23 May 2024 18:12:12 GMT From: Alexander Motin <mav@FreeBSD.org> To: src-committers@FreeBSD.org, dev-commits-src-all@FreeBSD.org, dev-commits-src-branches@FreeBSD.org Subject: git: 856d35337225 - releng/14.1 - Fix scn_queue races on very old pools Message-ID: <202405231812.44NICCwQ033048@gitrepo.freebsd.org>
next in thread | raw e-mail | index | archive | help
The branch releng/14.1 has been updated by mav: URL: https://cgit.FreeBSD.org/src/commit/?id=856d35337225d77948b43ee5d479baa2588963ec commit 856d35337225d77948b43ee5d479baa2588963ec Author: Alexander Motin <mav@FreeBSD.org> AuthorDate: 2024-05-23 16:20:37 +0000 Commit: Alexander Motin <mav@FreeBSD.org> CommitDate: 2024-05-23 18:11:36 +0000 Fix scn_queue races on very old pools Code for pools before version 11 uses dmu_objset_find_dp() to scan for children datasets/clones. It calls enqueue_clones_cb() and enqueue_cb() callbacks in parallel from multiple taskq threads. It ends up bad for scan_ds_queue_insert(), corrupting scn_queue AVL-tree. Fix it by introducing a mutex to protect those two scan_ds_queue_insert() calls. All other calls are done from the sync thread and so serialized. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Brian Atkinson <batkinson@lanl.gov> Signed-off-by: Alexander Motin <mav@FreeBSD.org> Sponsored by: iXsystems, Inc. Closes #16162 PR: 278414 Approved by: re (cperciva) (cherry picked from commit 49086aa35d987b78dbc3c9ec94814fe338e07164) (cherry picked from commit 455ce1729353f2ffce9713ccc3574e73186a22f0) --- sys/contrib/openzfs/include/sys/dsl_scan.h | 1 + sys/contrib/openzfs/module/zfs/dsl_scan.c | 6 ++++++ 2 files changed, 7 insertions(+) diff --git a/sys/contrib/openzfs/include/sys/dsl_scan.h b/sys/contrib/openzfs/include/sys/dsl_scan.h index 2e3452e5ebaa..f32f59a2bedf 100644 --- a/sys/contrib/openzfs/include/sys/dsl_scan.h +++ b/sys/contrib/openzfs/include/sys/dsl_scan.h @@ -173,6 +173,7 @@ typedef struct dsl_scan { dsl_scan_phys_t scn_phys; /* on disk representation of scan */ dsl_scan_phys_t scn_phys_cached; avl_tree_t scn_queue; /* queue of datasets to scan */ + kmutex_t scn_queue_lock; /* serializes scn_queue inserts */ uint64_t scn_queues_pending; /* outstanding data to issue */ /* members needed for syncing error scrub status to disk */ dsl_errorscrub_phys_t errorscrub_phys; diff --git a/sys/contrib/openzfs/module/zfs/dsl_scan.c b/sys/contrib/openzfs/module/zfs/dsl_scan.c index 34012db82dee..c509f402c44a 100644 --- a/sys/contrib/openzfs/module/zfs/dsl_scan.c +++ b/sys/contrib/openzfs/module/zfs/dsl_scan.c @@ -491,6 +491,7 @@ dsl_scan_init(dsl_pool_t *dp, uint64_t txg) avl_create(&scn->scn_queue, scan_ds_queue_compare, sizeof (scan_ds_t), offsetof(scan_ds_t, sds_node)); + mutex_init(&scn->scn_queue_lock, NULL, MUTEX_DEFAULT, NULL); avl_create(&scn->scn_prefetch_queue, scan_prefetch_queue_compare, sizeof (scan_prefetch_issue_ctx_t), offsetof(scan_prefetch_issue_ctx_t, spic_avl_node)); @@ -646,6 +647,7 @@ dsl_scan_fini(dsl_pool_t *dp) scan_ds_queue_clear(scn); avl_destroy(&scn->scn_queue); + mutex_destroy(&scn->scn_queue_lock); scan_ds_prefetch_queue_clear(scn); avl_destroy(&scn->scn_prefetch_queue); @@ -2727,8 +2729,10 @@ enqueue_clones_cb(dsl_pool_t *dp, dsl_dataset_t *hds, void *arg) return (err); ds = prev; } + mutex_enter(&scn->scn_queue_lock); scan_ds_queue_insert(scn, ds->ds_object, dsl_dataset_phys(ds)->ds_prev_snap_txg); + mutex_exit(&scn->scn_queue_lock); dsl_dataset_rele(ds, FTAG); return (0); } @@ -2919,8 +2923,10 @@ enqueue_cb(dsl_pool_t *dp, dsl_dataset_t *hds, void *arg) ds = prev; } + mutex_enter(&scn->scn_queue_lock); scan_ds_queue_insert(scn, ds->ds_object, dsl_dataset_phys(ds)->ds_prev_snap_txg); + mutex_exit(&scn->scn_queue_lock); dsl_dataset_rele(ds, FTAG); return (0); }
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?202405231812.44NICCwQ033048>